SignalFx Microservices APM™️ (µAPM) is the only solution that provides a directed approach to troubleshooting. Unlike traditional APM solutions, µAPM analyzes end-to-end traces across service boundaries with open instrumentation options to provide flexibility for the end-user to collect traces. With SignalFx’s NoSample™️ tail-based architecture, you’re able to capture every trace, and pinpoint trace anomalies in real-time. With deeper analysis than any other solution the SignalFx µAPM Outlier Analyzer™️ enables insights into the patterns of your spans in your anomalous traces.

Analyze Every Transaction, Focus on What Matters Most

SignalFx Smart Gateway observes every transaction across distributed services and uses machine learning algorithms to define a true anomaly.

  • Never miss an anomaly – Deliver flawless end-user experience by catching performance outliers across all the microservices;
  • Cut through the noise – Using statistical computations, Smart Gateway cuts through the noise and enables you to focus on the traces that matter;
  • Capture one-in-a-thousand issues – Traditional APM tools use random and probabilistic sampling resulting in missing important diagnostic data when it is needed the most – during troubleshooting. The SignalFx Smart Gateway intelligently captures every edge case, such as p(99) outliers, so you can confidently address intermittent issues.

Quickly Isolate the Root Cause

SignalFx provides a unified, single-pane-of-glass view of infrastructure and applications monitoring – all contextual and fully correlated.

  • Narrow down the root cause within seconds to infrastructure, PaaS or application code;
  • Correlate how infrastructure performance is impacting application performance;
  • Navigate to the application code-level insights within seconds using contextual alerts pointing to the relevant traces;
  • Reduce MTTR and enable DevOps practices by having a single source of truth across infrastructure and deployed microservices.

Understand Microservices Dependencies with out-of-the-box Service Maps

SignalFx instantly discovers all monitored microservices and their interdependencies in real-time.

  • Quickly visualize service health, data-flows and interactions using Service Maps;
  • Rapidly pinpoint the performance issues using the contextual service maps showing only the services that participated in a specific trace;
  • One-click navigation to p(99), edge-case or intermittent high latency issues using Latency Histogram.

Optimize with Contextual Insights

SignalFx converts span, trace and path level data into metrics to enable real-time analytics.

  • Release code with confidence by comparing service, trace or even span level performance metrics across canary deployments and multiple versions of each microservice;
  • Quickly find service performance related to customer segments, user ID or any custom tag;
  • Slice and dice performance data for service transaction paths, the logical representation of how your end-users interact with your applications.

Get Complete Flexibility with Open, Vendor-Neutral Instrumentation Standards.

Automatically collect trace data from the most popular open source frameworks and libraries. Vendor-neutral, open standards support avoids proprietary, heavy-weight APM agents and the production overhead

  • Get complete flexibility in instrumenting your code using OpenTracing, Zipkin, OpenCensus or transparently get trace data using service meshes such as Istio;
  • Native support for OpenTracing and hundreds of supported open source libraries and frameworks;
  • Seamless agent installation and automatically start sending trace data.

Glympse uses SignalFx Microservices APM monitoring

Glympse uses SignalFx analytics to predict and completely prevent outages

Glympse compares real-time customer visits against historical trends to quickly spot anomalous activity and potential issues. SignalFx dashboards act as a leading indicator showing when and where problems originate and providing 5-10 minutes early warning — helping Glympse prevent application outages completely.